During replication, HIV-1 converts its packaged dimeric RNA genomes into DNA and generates one provirus in an infection event. In this process, HIV-1 needs to preserve its genetic information while the host innate immune response attempts to abolish the generation of proviruses capable of producing infectious progeny. In collaboration with Dr. Vinay Pathak's section (HIV Dynamics and Replication Program), we examined whether HIV-1 genomes with hypermutation generated by host restriction APOBEC3 proteins could be rescued by recombination or only suffered sublethal mutations. Our experimental data and analyses demonstrated that both scenarios occurred rarely; therefore, hypermutation mostly results in dead-end viral genomes that do not contribute to the diversity of the replicating viral population. Additionally, we have assisted our collaborators in completing several studies on retroviral replication, including developing better reporters for cell-to-cell infection, following the HIV-1 complex in infected cells, characterizing the restriction factors APOBEC3G and APOBEC3F, and studying the precursors of xenotropic murine leukemia virus-related viruses. Our current and future research efforts are focused on examining whether recombination is a key mechanism to maintain HIV-1 genome integrity and defining the mechanism that causes HIV-1 recombination hot spots. These studies seek to understand how HIV-1 maintains its genome and reassorts mutations to increase viral diversity. __BACKGROUND: HIV-1 packages two copies of viral RNA into one virion, although each RNA contains all of the information required for viral replication. It has long been speculated that the selective advantages for packaging two copies of viral RNA are to rescue the genetic information from breaks in the RNAs and to allow recombination to assort mutations and increase viral diversity. Although often assumed, these two functions have not been tested experimentally; we will address these hypotheses in our research. Frequent retroviral recombination occurs during DNA synthesis when virally encoded RT uses a portion of each copackaged RNA as a template. The resulting DNA is mosaic in its genetic composition and contains portions of its sequence from each copackaged viral RNA molecule. Recombination can occur in all HIV-1 particles; however, progeny distinct from parents are generated from heterozygous particles that contain two different RNAs. Heterozygous particles are produced from cells coinfected with more than one virus and by copackaging RNAs derived from two different proviruses. Although recombination can occur in homozygous particles, which contain two copies of RNA derived from the same provirus, the resulting recombinants have the same genotype as the parent virus. Recombination has been observed to occur throughout the HIV-1 genome; however, recombination hot spots have been reported and are thought to be caused by RNA structures. We will examine the relationship between RNA structures and recombination hot spots. __ACCOMPLISHMENTS: Recombination occurs frequently during the synthesis of HIV-1 DNA. Although the role of recombination in generating viral diversity has been established, it is unknown whether recombination is required to complete DNA synthesis. We have investigated the role of recombination in viral replication by blocking recombination in a portion of the viral genome. We found that viral titer decreased when recombination is blocked. Furthermore, the remaining proviruses have high deletion rates. These studies indicate that recombination is a mechanism to salvage viral genetic information and maintain genome integrity. __Host restriction APOBEC3 (A3) proteins can induce G-to-A hypermutation. Currently, numerous studies have disagreed on the contribution of hypermutation to viral genetic diversity and evolution. In collaboration with the Pathak group, we determined the effects of hypermutation on the HIV-1 recombination rate and how often hypermutated sequences could be rescued via recombination to produce viable genomes and contribute to genetic variation. We found that hypermutation did not significantly affect the rate of recombination, and recombination between hypermutated and wild-type genomes only increased the viral mutation rate by 3.9 x 10-5 mutations/bp/replication cycle in heterozygous virions, which is similar to the HIV-1 mutation rate. Since copackaging of hypermutated and wild-type genomes occurs very rarely in vivo, recombination between hypermutated and wild-type genomes does not significantly contribute to the genetic variation of replicating HIV-1. We also analyzed previously reported hypermutated sequences from infected patients and determined that the frequency of sublethal mutagenesis for A3G and A3F is negligible (1 x 10-11) and its contribution to the viral mutations is far below that of mutations introduced by reverse transcription. Overall, we conclude that the contribution of A3-induced hypermutation to HIV-1 genetic variation is lower than the contribution of mutations during error-prone replication.